ISEI Visitors Need Support to Understand Current Scientific Research

advertisement
ISEI Visitors Need Support to Understand Current Scientific Research
Presented in Data Visualizations on 3-D Global Projection Systems
Kathryn Stofer – Oregon State University // stoferk@onid.orst.edu
Céleste Frazier Barthel – Oregon State University and Wilson College // cbarthel@wilson.edu
Shawn Rowe – Oregon State University and Oregon Sea Grant // shawn.rowe@oregonstate.edu
Little is known about how
informal science education
institution (ISEI) visitors make
meaning from these images.
These platforms were introduced to museums around 2004. Since then,
few research studies but some evaluations have been done. Visitors are
attracted to these exhibits but don’t stay long unless there is a docent-led
presentation or an interactive kiosk to provide explanations about the
images. At globes with strictly auto-run presentations, dwell time has a
median of less than a minute, while docent-led presentations have a dwell
time of nearly 18 minutes. Kiosk-driven globes have a median dwell time of
about 6 minutes.
Novices benefit from familiar color,
data scales and labels indicating
geography, date and time.
On their own, many scientific images are obscure to visitors (Phipps and
Rowe, 2010), and many content questions and statements in visitor groups
revolve around clarifying what the images are depicting (Rowe, et. al., 2010,
Barthel, 2010). Visitors are confused when the same color schemes are
used on several different images to mean different things (Rowe, et. al.,
2010; Phipps and Rowe, 2010; Barthel, 2010). Date and time are not easily
transmitted via accompanying text (Rowe, et. al., 2010).
Photo credit: Flickr – woodleywonderworks
We used a combination of
quantitative and qualitative
methods to explore interactions
in multiple settings.
Geographic
Labels
Familiar
Colors
Quantitative data collection included unobtrusive observations of visitors
recording dwell time and counting particular interactions. Qualitative data
collection included videotaping and coding interactions, as well as in-depth
interviews with visitors with or without stimulated recall. Data was collected
at one 6-ft. diameter globe with live or auto-run presentations at a large
independent science center on the East Coast, and at one 3-ft. globe with
kiosk-driven interactions at a university-operated science center on the West
Coast. A variety of images were used.
Familiar
Scale Units
Only experts could understand
this “scientific” image.
Photo credit: Rowe, et. al., 2010
Visitors do not automatically
recognize that the images and
animations are from
remote-sensing data.
Barthel (2010) found visitors 1) did not recognize the term “remote sensing”
until it was explained in the context of something such as GPS, 2) thought
the images were pictures of Earth, not visualizations of data, and 3) thought
the false colors were produced by camera lenses. Rowe, et. al. (2010) also
found confusion as to the sources of the images even with narration and
text in an accompanying kiosk.
Visitors use prior knowledge
and experiences to try to
make meaning as a group.
Constructivism suggests that meaning is made through putting
new knowledge in context with prior knowledge. Our results show
that visitors talk about prior knowledge and experiences but do so
in order to share with their groups, often using deictic words,
indefinite words or gestures that need context, to indicate
presumed or attempted shared meaning, a more sociocultural
approach to learning (Rowe, et. al., 2010, Barthel, 2010).
Both experts and novices recognize this depicts
ocean temperatures in the northeastern Pacific.
Visualizations cannot be simply
re-purposed as-is from scientific
publications for the public to
make meaning from them.
Visualizations of global Earth system science data must use significant
scaffolding for educational use. With support, visitors can be helped to
understand the scientific data represented. When visualizations are more readily
understandable, docents and/or kiosks can serve to highlight features relevant
to problems and thus support visitors in making judgments related to policy or
lifestyle decisions as science literate citizens.
Download